MiMo-V2.5-ASR MLX BF16

This repository contains the bfloat16 ASR checkpoint package generated from the local MiMo-V2.5-ASR safetensors. It is the current default precision for the mimo_mlx Apple Silicon inference code.

Files

  • model-00001-of-00007.safetensors ... model-00007-of-00007.safetensors
  • model.safetensors.index.json
  • Qwen tokenizer/config sidecar files

Load

from mimo_mlx import load_asr

asr = load_asr("bf16")
text = asr.transcribe("audio.wav", language="zh")

The audio tokenizer is required separately. Use the official XiaomiMiMo/MiMo-Audio-Tokenizer files or the local directory expected by the project.

Validation

On the local VoiceInk-History seeded spread check, bf16 preserved the validated float32 behavior.

Source

Base model: https://huggingface.co/XiaomiMiMo/MiMo-V2.5-ASR

Downloads last month
299
Safetensors
Model size
8B params
Tensor type
BF16
·
MLX
Hardware compatibility
Log In to add your hardware

Quantized

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for carloshuang1224/MiMo-V2.5-ASR-MLX-BF16

Finetuned
(4)
this model